Two microphone technique to improve the speech intelligibility under noisy environment

摘要

This contribution presents an efficient technique for the speech enhancement of a signal using Kepstrum based approach. In any speech enhancement system, the noise is being suppressed in spectral domain assuming the degraded component are stationary or slowly varying for few frames of the signal. As conventional methods for noise elimination fails due to non-stationary properties of the speech signal, in this context, a distortion filter and speech remover filter are being introduced. The main focus of this paper is involved in developing a new speech enhancement system based on speech characteristics and it's modelling provide high quality and intelligibility on mobiles or embedded platforms that have two or more microphones. The proposed method was evaluated for machinery, babble, and car noise types, using both objective and subjective measures for both quality and intelligibility of the enhanced speech. The results show significant improvements in comparison with single microphone speech enhancement, kepstrum and Distortion Correlation Filter methods.